Typically, yes. We code things up independently and then verify that our results match. I tend to do this especially with early student/new co-authors since we are more likely to disagree on choices we make. Over time (across multiple projects) as I gain trust (and I do not trust easily) I agree that I do not go over the code with a fine tooth comb, but I always go through the modeling choices, the data cleaning process and run the final code to obtain results. Without this I am not sure how I put my name on the paper if I don't even know what is going on...
>> Senior faculty here. If your name is on the paper you are responsible for the content. End of story.
Absolutely.Yes but do YOU run the code? You’re putting your name on the line, absolutely, but you trust your juniors not to be fabricating data.